TomViolenz
2015-12-06 12:43:27
Hey guys
Here is a very satisfied customer, whose Bomes templates reach a level where the Midi processing seems to get a bit flaky at times. I'm running two different instances of Bomes MT Pro at the same time, each with several presets and hundreds of translators. Controlling 7 (soon 8 ) different controller. So this might be expected. Nonetheless I would like to know of some optimization strategies to make my Live set more reliable.
1: First thing I'm wondering is, if I'm really profiting from running two instances at the same time, which I assumed spreads out the processing better and makes Bomes indirectly a multicore application. But, since this set-up causes its own set of problems by doubling the existing Bomes Midi ports of the same name and adding a "2" after it, which does confuse Ableton Live at times, I would prefer not to do this further, if it doesn't benefit me.
(And as a slight add on question, would using port aliases get rid of this problem of Live confusing the ports? I somehow never quite understood their purpose)
2: Secondly, two of my controllers are 64 buttons matrixes (APC Minis), and programming these buttons to do what I want uses up LOTS of translators, which if I understand correctly slows down processing quite a bit. I could conceivably subsume a few of these by filtering the incoming Midi in each translator, which would more or less half the translator count in one preset from 150 to about 80. Would this make enough of a difference to make this rewriting effort worth the while? Especially considering that only a third of these translators actually receive Midi. (While the others get triggered as a result via timers).
So I'm basically wondering if the impact of the filtering, writing the CC numbers to variables, so that the one-shot timers know what CC needs to be processed etc. and the rather complicated translators I would get using "Goto-->Label" combinations, would still be less than having more (albeit simpler) translators.
3: I have one preset which I assume slows down the general processing quite a bit, since it constantly has to process Midi coming from a Midi envelope in a running clip in Live, filter it and start timers and then output Midi based on the resulting calculations. (This is the preset that I mostly run in the second instance of Bomes, so that it does not interfere with the functions of the rest of the translators). Is this something that could really bring Bomes to its knees, or am I being overly cautious? (I'm running this on a Quad Core i7 MBPro)
Any input to stabilize this set up would be greatly appreciated
One more, unrelated question: Is proper undo in the cards for Bomes? I can't count the number of times anymore that I lost lots of work, because of a stupid mistake or a fat finger, that I couldn't revert.
Here is a very satisfied customer, whose Bomes templates reach a level where the Midi processing seems to get a bit flaky at times. I'm running two different instances of Bomes MT Pro at the same time, each with several presets and hundreds of translators. Controlling 7 (soon 8 ) different controller. So this might be expected. Nonetheless I would like to know of some optimization strategies to make my Live set more reliable.
1: First thing I'm wondering is, if I'm really profiting from running two instances at the same time, which I assumed spreads out the processing better and makes Bomes indirectly a multicore application. But, since this set-up causes its own set of problems by doubling the existing Bomes Midi ports of the same name and adding a "2" after it, which does confuse Ableton Live at times, I would prefer not to do this further, if it doesn't benefit me.
(And as a slight add on question, would using port aliases get rid of this problem of Live confusing the ports? I somehow never quite understood their purpose)
2: Secondly, two of my controllers are 64 buttons matrixes (APC Minis), and programming these buttons to do what I want uses up LOTS of translators, which if I understand correctly slows down processing quite a bit. I could conceivably subsume a few of these by filtering the incoming Midi in each translator, which would more or less half the translator count in one preset from 150 to about 80. Would this make enough of a difference to make this rewriting effort worth the while? Especially considering that only a third of these translators actually receive Midi. (While the others get triggered as a result via timers).
So I'm basically wondering if the impact of the filtering, writing the CC numbers to variables, so that the one-shot timers know what CC needs to be processed etc. and the rather complicated translators I would get using "Goto-->Label" combinations, would still be less than having more (albeit simpler) translators.
3: I have one preset which I assume slows down the general processing quite a bit, since it constantly has to process Midi coming from a Midi envelope in a running clip in Live, filter it and start timers and then output Midi based on the resulting calculations. (This is the preset that I mostly run in the second instance of Bomes, so that it does not interfere with the functions of the rest of the translators). Is this something that could really bring Bomes to its knees, or am I being overly cautious? (I'm running this on a Quad Core i7 MBPro)
Any input to stabilize this set up would be greatly appreciated
One more, unrelated question: Is proper undo in the cards for Bomes? I can't count the number of times anymore that I lost lots of work, because of a stupid mistake or a fat finger, that I couldn't revert.